Intelligent bus timetable dynamic generation method and system based on multi-source data

Abstract

The invention discloses an intelligent bus timetable dynamic generation method and system based on multi-source data, and belongs to the technical field of intelligent city buses, and the method comprises the following steps: 1, analyzing a city bus multi-line data file, and obtaining a data set of line station names, longitudes and latitudes, historical passenger flow data, initial departure time and legal departure frequency; 2, time periods are dynamically divided based on a dynamic time period division algorithm and passenger flow, an optimization model is constructed by adopting three timetable generation principles and four departure frequency calculation methods, and an optimized departure time sequence of each station is generated based on the initial departure time and the dynamic optimization model; 3, adjusting the departure-arrival time of multiple routes at the transfer station through a transfer cooperation algorithm; and step 4, outputting the optimized departure time table and a visual output result. The problem of low efficiency of traditional scheduling can be solved, the operation cost is reduced, and the satisfaction degree of passengers is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent city public transport technology, and particularly relates to a dynamic generation method and system of intelligent public transport timetable based on multi-source data, which is suitable for scheduling optimization and operation management of city public transport enterprises. BACKGROUND

[0002] As an important part of urban public transport, the operation efficiency and service quality of city public transport system directly affect the travel experience of citizens and the traffic condition of the city. The traditional public transport diagram mainly relies on manual drawing and experience judgment, and has problems such as low efficiency, difficulty in coping with complex road conditions and unexpected situations. The present application realizes the intelligentization and refinement of public transport scheduling through a data-driven dynamic optimization model and interactive visualization, and can significantly improve the operation efficiency of the public transport system and further optimize the travel experience of passengers through visual decision-making assistance. SUMMARY

[0003] The purpose of the present application is to provide a dynamic generation method and system of intelligent public transport timetable based on multi-source data, which solves the problem of low efficiency of traditional scheduling under the condition of balanced service of public transport vehicles in transition period, reduces operation cost and improves passenger satisfaction.

[0004] To achieve the above purpose, the present application provides a dynamic generation method and system of intelligent public transport timetable based on multi-source data, which comprises:

[0005] A dynamic generation method of intelligent public transport timetable based on multi-source data, comprising the following steps:

[0006] Step 1: Analyzing city public transport multi-line data files to obtain data sets of line station name, longitude and latitude, historical passenger flow data, initial departure time and legal departure frequency;

[0007] Step 2: Based on a dynamic period division algorithm and passenger flow dynamic division period T j , three kinds of timetable generation principles and four kinds of departure frequency calculation methods are used to construct an optimization model, and based on the initial departure time and the dynamic optimization model, the optimized departure time sequence of each station is generated;

[0008] Step 3: Adjusting the departure-arrival time of multi-line at transfer stations through a transfer coordination algorithm;

[0009] Step 4: Outputting the optimized departure timetable and visual output result.

[0010] Preferably, in step 1, the multi-line data file comprises:

[0011] The number of stations N 站 of each line, running speed v k, vehicle capacity c, legal departure frequency F 法定 , departure time HH:MM, selection principle and method, uniform passenger load d oj , adjacent station spacing d i,i+1 , historical passenger flow data P hist (t), and station latitude and longitude (x i , y i ).

[0012] Preferably, in step 2, the dynamic time period division algorithm is:

[0013] T j ={t∣μ-1.5σ≤P(t)≤μ+1.5σ};

[0014] In the formula, T j is the passenger flow dynamic division period, indicating the time period divided according to passenger flow fluctuation; t is the absolute time, indicating the specific time point (hour or minute) in a day; μ is the daily average passenger flow of historical passenger flow data; σ is the standard deviation of historical passenger flow data, reflecting the passenger flow fluctuation amplitude; P(t) is the real-time passenger flow at time t;

[0015]

[0016] In the formula, P(h) is the passenger flow of the hth hour;

[0017]

[0018] The dynamic time period division algorithm includes holiday correction:

[0019]

[0020] In the formula, is the dynamic time period after holiday correction; μ h is the average passenger flow during holidays; σ h is the standard deviation of passenger flow during holidays;

[0021] Three kinds of timetable generation principles are defined as:

[0022] Smooth transition uniform departure interval M1 principle:

[0023]

[0024] In the formula, t n is the departure time of the nth train; t n-1 is the departure time of the (n-1)th train; T 运营 is the total daily operation time of the line; N is the total number of daily departures; n is the current sequence number, 1≤n≤N; α is the maximum adjustment amplitude, α=0.2;

[0025] Average passenger load balancing M2 principle:

[0026]

[0027] where P(τ) is the real-time passenger flow at time τ; τ is the relative time; D 总 is the total demand passenger flow in the time period;

[0028] Maximum single vehicle passenger load balancing M3 principle:

[0029]

[0030] where Q i (t) is the cumulative passenger flow of station i at time t; d oj is the uniform passenger load; c is the vehicle capacity;

[0031]

[0032] where P i (τ) is the passenger flow of station i at time τ;

[0033] The four kinds of departure frequency calculation methods are:

[0034] Daily maximum passenger flow method F1:

[0035]

[0036] where F 1j is the calculated departure frequency of period j using F1; P maxday is the maximum daily passenger flow of station i in period j; F 法定 is the legal departure frequency of the line;

[0037] Hourly maximum passenger flow method F2:

[0038]

[0039] where F 2j is the calculated departure frequency of period j using F2; P ij is the maximum hourly passenger flow of station i in period j;

[0040] Passenger kilometer method F3:

[0041]

[0042] where F 3j is the calculated departure frequency of period j using F3; A j is the total passenger kilometer number of period j; L is the total length of the line, in kilometers;

[0043]

[0044] where d i,i+1 is the distance between adjacent stations; is the average passenger flow of station i in period j;

[0045] Crowding constraint method F4:

[0046]

[0047] where F 4j is the calculated departure frequency of period j using F4; I j is the set of crowded stations;

[0048]

[0049] where i is the station number on the line.

[0050] Preferably, in step 3, the transfer coordination algorithm comprises:

[0051] Transfer correlation calculation:

[0052]

[0053] where R m,n is the correlation of lines m and n at transfer stations; S 换乘 is the set of transfer stations; P m,i is the passenger flow of line m at transfer station i; P n,i is the passenger flow of line n at transfer station i;

[0054] Departure time adjustment amount:

[0055]

[0056] where ΔT m,n is the departure time adjustment amount of lines m and n, in minutes; is the departure time of line n at transfer station i; is the arrival time of line m at transfer station i;

[0057] Optimization objective function:

[0058] minΣ m,n R m,n ·(ΔT m,n ) 2 ;

[0059] Transfer station capacity constraint:

[0060]

[0061] where Ki C is the set of lines passing through transfer station i; C i F is the platform capacity; F k P is the departure frequency of line k; P trans,k,i Q is the transfer passenger flow of line k at transfer station i; Q

[0062] Real-time dynamic adjustment:

[0063]

[0064] wherein P0 is the passenger flow threshold value; β is the response speed parameter, β = 0.05; F j (t) is the dynamic departure frequency of time period j at time t.

[0065] Preferably, in step 4, the output optimized departure schedule satisfies:

[0066]

[0067] wherein δ is the allowable deviation rate, δ = 0.1; F avg M is the total number of time periods; P is the departure frequency generated by the model; P is the actual required departure frequency.

[0068] Preferably, the weather influence factor:

[0069]

[0070] wherein R is the real-time rainfall, unit: mm / h; R0 is the reference value, R0 = 5; ξ is the sensitivity coefficient, ξ = 0.15; P is the corrected departure frequency (considering rainfall R).

[0071] Preferably, the energy consumption optimization constraint:

[0072]

[0073] wherein a is the basic energy consumption coefficient of unit departure frequency; b is the energy consumption coefficient of the square term of departure frequency; F 标 E is the rated departure frequency; E 总 E is the total energy consumption; E max E is the allowable maximum energy consumption threshold value.

[0074] Preferably, the passenger satisfaction evaluation model:

[0075]

[0076] wherein w i is the waiting time of the ith passenger; τ swhere is the satisfaction decay constant, τ s = 10; S is the overall satisfaction score, ranging from 0 to 1.

[0077] A multi-source data-based intelligent public transport timetable dynamic generation system, comprising the following modules:

[0078] Data processing module: parse Excel data file, execute dynamic time period division algorithm;

[0079] Model calculation module: execute M1-M3 principle, F1-F4 method and transfer coordination algorithm;

[0080] Dynamic correction module: real-time optimization combined with dynamic factors such as weather and passenger flow;

[0081] Output processing module: generate timetable file and visual output result.

[0082] Preferably, the data processing module includes an anomaly detection submodule:

[0083]

[0084] Trigger data cleaning when the anomaly score > 1;

[0085] In the formula, κ is the acceleration weight coefficient, κ = 0.3; is the average value of the distance between adjacent stations; σ d is the standard deviation of the distance between adjacent stations; P i is the real-time passenger flow of station i; is the historical average passenger flow of station i; σ P is the standard deviation of the passenger flow; P is the time series function P(t) of the passenger flow data.

[0086] Therefore, the application adopts the above-mentioned multi-source data-based intelligent public transport timetable dynamic generation method and system, which has the following beneficial effects compared with the prior art:

[0087] 1. Improve operational efficiency and reduce resource waste

[0088] Dynamic response to passenger flow changes: through multi-source data fusion and real-time analysis, the system can identify passenger flow peaks and flat peaks, dynamically adjust the departure frequency and the number of trips, and reduce the problems of vehicle empty running or excessive congestion caused by traditional fixed timetables.

[0089] Optimize vehicle scheduling: combine line running speed, station spacing and other parameters to scientifically allocate transport resources, reduce invalid mileage and energy consumption, and improve the overall utilization rate of public transport vehicles.

[0090] 2. Improve passenger travel experience

[0091] Shorten waiting time: Through passenger flow prediction and dynamic time period division, the system increases the departure frequency during peak hours and reasonably lengthens the departure interval during off-peak hours, making the waiting time of passengers more in line with actual needs.

[0092] Seamless transfer coordination: Based on the correlation analysis and time adjustment algorithm of transfer stations, the system reduces the waiting time of passengers during transfer and improves the convenience of multi-line joint operation.

[0093] 3. Enhance the intelligence and flexibility of scheduling decisions

[0094] Multi-dimensional data-driven: Integrate historical passenger flow, real-time passenger flow data, weather information and other multi-source data to support scheduling decisions and reduce the blindness of manual intervention.

[0095] Dynamic correction mechanism: Through real-time optimization modules such as anomaly detection and weather influence factors, the system can quickly respond to sudden situations (extreme weather or temporary activities) and ensure the robustness of the timetable.

[0096] 4. Support visualization and collaborative management

[0097] Intuitive visual output: Display key indicators such as departure frequency and passenger flow distribution through heat maps and Gantt charts to help dispatchers quickly grasp the overall operation status.

[0098] Interactive adjustment function: Allows managers to adjust the timetable generated by the system according to actual conditions, realizing efficient decision-making through human-machine collaboration.

[0099] 5. Promote green and low-carbon operation

[0100] Energy consumption optimization constraint: On the premise of meeting passenger flow demand, control the total departure frequency and vehicle configuration through algorithms to reduce unnecessary energy consumption and promote the sustainable development of the public transportation system.

[0101] The technical solutions of the present application will be further described in detail through the accompanying drawings and examples. DETAILED DESCRIPTION

[0102] Figure 1 The method flowchart of the present application;

[0103] Figure 2 The system structure diagram of the present application;

[0104] Figure 3 The xlsx data file of five bus lines in Hohhot City that needs to be entered in the present application;

[0105] Figure 4 The multi-line map visualization interface after data entry in the present application;

[0106] Figure 5 For the multi-line map visualization interface after hiding the site in the present application;

[0107] Figure 6 For the multi-line departure frequency heat map in the present application;

[0108] Figure 7 For the multi-line bus schedule Gantt chart in the present application;

[0109] Figure 8 For the multi-line passenger flow intensity distribution diagram in the present application;

[0110] Figure 9 For the generated departure schedule in the present application;

[0111] Figure 10 For the downloaded example file in the present application;

[0112] Figure 11 For the downloaded bus schedule file in the present application. DETAILED DESCRIPTION

[0113] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application are further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. The examples of the embodiments are shown in the drawings, wherein the same or similar reference numerals represent the same or similar elements or elements having the same or similar functions throughout.

[0114] It should be noted that the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0115] Similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0116] In the description of the present application, it should be noted that the terms "upper", "lower", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, or the orientation or positional relationship commonly used when the product of the present application is used, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0117] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "arrangement", "installation", "connection" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0118] Embodiment

[0119] As Figure 1 shown, the intelligent public transport timetable dynamic generation method based on multi-source data of the present application comprises the following steps:

[0120] Step 1: Analyze the city public transport multi-line data file to obtain the data set of line station name, latitude and longitude, historical passenger flow data, initial departure time and legal departure frequency.

[0121] In step 1, the multi-line data file comprises:

[0122] The number of stations N of each line 站 , running speed v k , vehicle capacity c, legal departure frequency F 法定 , departure time HH:MM, selected principle and method, uniform passenger load d oj , distance between adjacent stations d i,i+1 , historical passenger flow data P hist (t), and station latitude and longitude (x i , y i ).

[0123] Step 2: Based on the dynamic time period division algorithm and the passenger flow dynamic division time period T j , three kinds of timetable generation principles and four kinds of departure frequency calculation methods are used to construct an optimization model, based on the initial departure time and the dynamic optimization model, to generate the optimized departure time sequence of each station.

[0124] In step 2, the dynamic time period division algorithm is:

[0125] Tj ={t|μ-1.5σ≤P(t)≤μ+1.5σ};

[0126] In the formula, T j The dynamic passenger flow is divided into time periods, representing the time periods based on passenger flow fluctuations; t is the absolute time (departure time, time period division), representing a specific point in time (hour or minute) of the day; μ is the daily average passenger flow of historical passenger flow data; σ is the standard deviation of historical passenger flow data, reflecting the amplitude of passenger flow fluctuations; P(t) is the real-time passenger flow at time t.

[0127]

[0128] In the formula, P(h) represents the passenger flow in the h-th hour.

[0129]

[0130] The dynamic time period segmentation algorithm includes holiday correction:

[0131]

[0132] In the formula, This is a dynamic time period adjusted for holidays; μ h σ represents the average passenger flow during holidays; h This represents the standard deviation of passenger flow during holidays.

[0133] The three timetable generation principles are defined as follows:

[0134] The principle of smooth transition and uniform departure interval M1:

[0135]

[0136] In the formula, t n t represents the departure time of the nth bus. n-1 T represents the departure time of the (n-1)th bus. 运营 N represents the total daily operating time of the route; N represents the total number of daily departures; n represents the current departure number, 1≤n≤N; α represents the maximum adjustment range, α=0.2.

[0137] The principle of average passenger load balancing M2:

[0138]

[0139] In the formula, P(τ) represents the real-time passenger flow at time τ; τ represents the relative time (integral / cumulative variable); D 总 This represents the total demand for passenger flow during the specified time period.

[0140] The principle of M3, which balances the maximum passenger capacity of a single vehicle:

[0141]

[0142] In the formula, Q i (t) represents the cumulative passenger flow of station i at time t; d oj c represents the uniform passenger capacity; c represents the vehicle capacity.

[0143]

[0144] In the formula, P i (τ) represents the passenger flow at station i at time τ.

[0145] The four methods for calculating departure frequencies are as follows:

[0146] Daily maximum passenger flow method F1:

[0147]

[0148] In the formula, F 1j The departure frequency is calculated using F1 for time period j; P maxday F represents the daily passenger flow at the station with the highest passenger flow within time period j; 法定 This refers to the legal departure frequency for this route.

[0149] Maximum passenger flow method F2 per hour:

[0150]

[0151] In the formula, F 2j The departure frequency is calculated using F2 for time period j; P ij The maximum hourly passenger flow of station i within time period j.

[0152] Passenger Kilometers Law F3:

[0153]

[0154] In the formula, F 3j For time period j, the departure frequency is calculated using F3; A j is the total passenger-carrying kilometers for time period j; L is the total length of the route, in kilometers.

[0155]

[0156] In the formula, d i,i+1 The distance between adjacent stations; Let be the average passenger flow at station i within time period j.

[0157] Crowding Constraint Method F4:

[0158]

[0159] In the formula, F 4jThe departure frequency is calculated using F4 for time period j; I j A collection of congested stations.

[0160]

[0161] In the formula, i is the station number on the line.

[0162] Step 3: Adjust the departure-arrival times of multiple routes at transfer stations using a transfer coordination algorithm.

[0163] In step 3, the transfer coordination algorithm includes:

[0164] Transfer correlation calculation:

[0165]

[0166] In the formula, R m,n S represents the correlation between lines m and n at the transfer station; 换乘 Gathering at the transfer station; P m,i P represents the passenger flow of line m at transfer station i; n,i Let be the passenger flow of line n at transfer station i.

[0167] Departure time adjustment amount:

[0168]

[0169] In the formula, ΔT m,n The adjustment amount for departure times of routes m and n, in minutes; Let n be the departure time of line n at transfer station i; Let be the arrival time of line m at transfer station i.

[0170] Optimize the objective function:

[0171] minΣ m,n R m,n ·(ΔT m,n ) 2 .

[0172] Transfer station capacity constraints:

[0173]

[0174] In the formula, K i C is the set of lines that pass through transfer station i; i For platform capacity; F k P is the departure frequency of line k; trans,k,i Let be the passenger flow at transfer station i for line k.

[0175] Real-time dynamic adjustment:

[0176]

[0177] In the formula, P0 is the passenger flow threshold; β is the response speed parameter, β = 0.05; F j (t) represents the dynamic departure frequency of time period j at time t.

[0178] Step 4: Output the optimized departure timetable and visualization results.

[0179] In step 4, the optimized departure timetable is output to satisfy:

[0180]

[0181] In the formula, δ is the allowable deviation rate, δ = 0.1; F avg M represents the average departure frequency; M represents the total number of time periods. The departure frequency generated for the model; The departure frequency is based on actual demand.

[0182] In addition, weather influencing factors, energy optimization constraints, and passenger satisfaction evaluation models are also considered.

[0183] Weather influencing factors:

[0184]

[0185] In the formula, R is the real-time rainfall, in mm / h; R0 is the baseline value, R0 = 5; ξ is the sensitivity coefficient, ξ = 0.15; This is the corrected departure frequency (taking into account rainfall R).

[0186] Energy consumption optimization constraints:

[0187]

[0188] In the formula, a is the basic energy consumption coefficient per unit departure frequency; b is the energy consumption coefficient of the square term of departure frequency; F 标 This refers to the rated departure frequency; E 总 Total energy consumption; E max This is the maximum allowable energy consumption threshold.

[0189] Passenger satisfaction evaluation model:

[0190]

[0191] In the formula, w i τ is the waiting time for the i-th passenger; s Let τ be the satisfaction decay constant per minute. s =10; S is the overall satisfaction score, ranging from 0 to 1.

[0192] The intelligent bus timetable dynamic generation system based on multi-source data described in this invention includes the following modules:

[0193] Data processing module: Parses Excel data files and executes dynamic time period segmentation algorithms.

[0194] Model calculation module: executes the M1-M3 principle, F1-F4 method and transfer coordination algorithm.

[0195] Dynamic correction module: Real-time optimization based on dynamic factors such as weather and passenger flow.

[0196] Output processing module: Generates timetable files and visualizes the output results.

[0197] The data processing module includes an anomaly detection submodule:

[0198]

[0199] Data cleaning is triggered when the outlier score is greater than 1.

[0200] In the formula, κ is the acceleration weighting coefficient, κ = 0.3; d is the average distance between adjacent stations; σ d P represents the standard deviation of the distance between adjacent stations. i Real-time visitor traffic for site i; σ represents the historical average passenger flow of station i; P Let P be the standard deviation of passenger flow; P is the time series function of passenger flow data, P(t).

[0201] like Figure 2 As shown, the system structure diagram of the present invention is explained as follows:

[0202] 1. Data Input and Processing

[0203] Users can refer to the required file format and information elements by clicking "Download Sample File" on the system's interactive interface. They then create a CSV or XLSX format line data file to be uploaded, select the corresponding data file in "Select File," and click "Upload Data File" to submit the file. The data processing module automatically identifies outliers (such as negative station spacing) and prompts the user to correct them. Figure 3 This demonstration uses actual passenger flow data from five bus routes in Hohhot from October 20, 2024 to November 3, 2024. Figure 10 This is a sample file in .xlsx format that you obtained.

[0204] 2. Solving the time-segmented model

[0205] The system dynamically divides time periods, generates departure intervals for each time period based on user-selected principles and methods, and combines transfer coordination algorithms, ultimately forming a departure timetable for the entered routes within the planned time periods. Figure 9 The departure schedule is derived from the actual passenger flow data of five bus routes in Hohhot from October 20, 2024 to November 3, 2024.

[0206] 3. Visual rendering

[0207] The map uses Leaflet.js, which combines user-uploaded station latitude and longitude coordinates with GeoJSON data for each route to form the route map showing the route direction and station locations. It renders a heatmap of departure frequency, a Gantt chart of bus schedules, and a passenger flow intensity distribution map for different time periods, which can be selectively viewed via the sidebar. Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 This is a visualization output generated based on the actual passenger flow data of five bus routes in Hohhot from October 20, 2024 to November 3, 2024.

[0208] 4. Export bus timetable

[0209] Click "Export Timetable" in the sidebar to automatically download the bus timetable file in xlsx format. Figure 11 This is an .xlsx file containing departure timetables obtained from actual passenger flow data of five bus routes in Hohhot from October 20, 2024 to November 3, 2024.

[0210] The scope of protection of this invention includes:

[0211] 1. Dynamic time period segmentation method for multi-source urban public transport data: Based on the statistical characteristics of historical passenger flow data, time periods are segmented by mean and standard deviation, and a holiday correction mechanism is included.

[0212] 2. Principles and methods for generating time-segmented bus timetables:

[0213] Three principles for generating departure times: smooth transition and uniform departure interval, balanced average passenger capacity, and balanced maximum passenger capacity per vehicle.

[0214] Four methods for calculating departure frequency: frequency calculation model based on daily maximum passenger flow, hourly maximum passenger flow, passenger-carrying kilometers and congestion constraints.

[0215] 3. Multi-line transfer collaborative optimization algorithm: a departure time adjustment model based on the passenger flow correlation of transfer stations, combined with a multi-line collaborative scheduling method with platform capacity constraints.

[0216] 4. Real-time correction and constraint mechanism for bus timetables:

[0217] A departure frequency correction method considering weather influencing factors; a dynamic adjustment algorithm based on real-time passenger flow fluctuations; and an energy consumption optimization constraint and timetable generation deviation control method.

[0218] 5. Passenger Satisfaction Evaluation Model: A quantitative evaluation method for passenger satisfaction based on passenger waiting time.

[0219] 6. Public transport data anomaly detection and cleaning technology: combining anomaly score calculation and cleaning triggering mechanism based on station spacing, passenger flow data and change rate.

[0220] 7. Intelligent bus timetable dynamic generation system based on the above method: system architecture including data processing, model calculation, transfer optimization and output modules.

[0221] 8. Map-based multi-line operation status visualization technology and interactive user interface design that supports parameter adjustment and data export.

[0222] 9. Application of the above technologies in intelligent bus dispatching and related improvement schemes.

[0223] Therefore, the present invention adopts the above-mentioned intelligent bus timetable dynamic generation method and system based on multi-source data, which solves the inefficiency problem of traditional scheduling, reduces operating costs, and improves passenger satisfaction under the condition of balanced bus service during transition periods.

[0224] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, this...

[0225] Those skilled in the art should understand that they can still make modifications to the technical solutions of this invention.

[0226] Modifications or equivalent substitutions are made, but these modifications or equivalent substitutions cannot improve the modified technology.

[0227] The proposed solution deviates from the spirit and scope of the technical solution of this invention.

Claims

1. A method for dynamically generating intelligent bus timetables based on multi-source data, characterized in that: Includes the following steps: Step 1: Parse the city bus multi-route data file to obtain the dataset containing route and station names, latitude and longitude coordinates, historical passenger flow data, initial departure time, and legal departure frequency; Step 2: Based on the dynamic time period segmentation algorithm and the dynamic passenger flow segmentation time period T j An optimization model was constructed using three timetable generation principles and four departure frequency calculation methods. Based on the initial departure time and the dynamic optimization model, the optimized departure time series for each station was generated. Step 3: Adjust the departure-arrival times of multiple routes at transfer stations using a transfer coordination algorithm; Step 4: Output the optimized departure timetable and visualization results.

2. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 1, characterized in that: In step 1, the multi-line data file includes: The number of stations N on each line 站 Operating speed v k Vehicle capacity (c) and legal departure frequency (F) 法定 Departure time (HH:MM), selection principle and method, uniform passenger capacity (d) oj Inter-station spacing d between adjacent stations i,i+1 Historical passenger flow data P hist (t), and the station's latitude and longitude (x) i ,y i ).

3. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 2, characterized in that: In step 2, the dynamic time period segmentation algorithm is as follows: T j ={t∣μ-1.5σ≤P(t)≤μ+1.5σ}; In the formula, T j The passenger flow is dynamically divided into time periods; t is the absolute time; μ is the daily average passenger flow of historical passenger flow data; σ is the standard deviation of historical passenger flow data; P(t) is the real-time passenger flow at time t; In the formula, P(h) represents the passenger flow in the h-th hour; The dynamic time period segmentation algorithm includes holiday correction: In the formula, This is a dynamic time period adjusted for holidays; μ h σ represents the average passenger flow during holidays; h This represents the standard deviation of passenger flow during holidays; The three timetable generation principles are defined as follows: The principle of smooth transition and uniform departure interval M1: In the formula, t n t represents the departure time of the nth bus. n-1 T represents the departure time of the (n-1)th bus. 运营 N represents the total daily operating time of the route; N represents the total number of daily departures; n represents the current departure number, 1≤n≤N; α represents the maximum adjustment range, α=0.2; The principle of average passenger load balancing M2: In the formula, P(τ) represents the real-time passenger flow at time τ; τ represents the relative time; D 总 This represents the total demand for passenger traffic within a given time period. The principle of M3, which balances the maximum passenger capacity of a single vehicle: In the formula, Q i (t) represents the cumulative passenger flow of station i at time t; d oj c represents the uniform passenger capacity; c represents the vehicle's carrying capacity. In the formula, P i (τ) represents the passenger flow at station i at time τ; The four methods for calculating departure frequencies are as follows: Daily maximum passenger flow method F1: In the formula, F 1j The departure frequency is calculated using F1 for time period j; P maxday F represents the daily passenger flow at the station with the highest passenger flow within time period j; 法定 This refers to the legally mandated departure frequency for this route; Maximum passenger flow method F2 per hour: In the formula, F 2j The departure frequency is calculated using F2 for time period j; P ij The maximum hourly passenger flow of station i within time period j; Passenger Kilometers Law F3: In the formula, F 3j For time period j, the departure frequency is calculated using F3; A j L represents the total passenger-carrying kilometers during time period j; L represents the total length of the route. In the formula, d i,i+1 The distance between adjacent stations; The average passenger flow at station i within time period j; Crowding Constraint Method F4: In the formula, F 4j The departure frequency is calculated using F4 for time period j; I j For congested stations; In the formula, i is the station number on the line.

4. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 3, characterized in that: In step 3, the transfer coordination algorithm includes: Transfer correlation calculation: In the formula, R m,n S represents the correlation between lines m and n at the transfer station; 换乘 Gathering at the transfer station; P m,i P represents the passenger flow of line m at transfer station i; n,i Let n be the passenger flow at transfer station i; Departure time adjustment amount: In the formula, ΔT m,n The adjustment amount for the departure time of routes m and n; Let n be the departure time of line n at transfer station i; Let m be the arrival time of line m at transfer station i; Optimize the objective function: min∑ m,n R m,n ·(ΔT m,n ) 2 ; Transfer station capacity constraints: In the formula, K i C is the set of lines that pass through transfer station i; i For platform capacity; F k P is the departure frequency of line k; trans,k,i Let K be the passenger flow at transfer station i for line k. Real-time dynamic adjustment: In the formula, P0 is the passenger flow threshold; β is the response speed parameter, β = 0.05; F j (t) represents the dynamic departure frequency of time period j at time t.

5. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 4, characterized in that: In step 4, the optimized departure timetable is output to satisfy: In the formula, δ is the allowable deviation rate, δ = 0.1; F avg M represents the average departure frequency; M represents the total number of time periods. The departure frequency generated for the model; The departure frequency is based on actual demand.

6. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 5, characterized in that: Weather influencing factors: In the formula, R is the real-time rainfall, in mm / h; R0 is the baseline value, R0 = 5; ξ is the sensitivity coefficient, ξ = 0.15; This is the corrected departure frequency.

7. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 6, characterized in that: Energy consumption optimization constraints: In the formula, a is the basic energy consumption coefficient per unit departure frequency; b is the energy consumption coefficient of the square term of departure frequency; F 标 This refers to the rated departure frequency; E 总 Total energy consumption; E max This is the maximum allowable energy consumption threshold.

8. The method for dynamically generating intelligent bus timetables based on multi-source data according to claim 7, characterized in that: Passenger satisfaction evaluation model: In the formula, w i τ is the waiting time for the i-th passenger; s Let τ be the satisfaction decay constant per minute. s =10; S is the overall satisfaction score.

9. A dynamic intelligent bus timetable generation system based on multi-source data, employing the dynamic intelligent bus timetable generation method based on multi-source data as described in any one of claims 1-8, characterized in that: Includes the following modules: Data processing module: Parses Excel data files and executes dynamic time period segmentation algorithms; Model calculation module: executes the M1-M3 principle, F1-F4 method and transfer coordination algorithm; Dynamic correction module: performs real-time optimization by combining dynamic factors such as weather and passenger flow; Output processing module: Generates timetable files and visualizes the output results.

10. The intelligent bus timetable dynamic generation system based on multi-source data according to claim 9, characterized in that: The data processing module includes an anomaly detection submodule: Data cleaning is triggered when the outlier score is greater than 1. In the formula, κ is the acceleration weighting coefficient, κ=0.3; σ is the average distance between adjacent stations; d P represents the standard deviation of the distance between adjacent stations. i Real-time visitor traffic for site i; σ represents the historical average passenger flow of station i; P Let P be the standard deviation of passenger flow; P is the time series function of passenger flow data, P(t).

Images